Childhood Onset Disorders: Applying the T/C Model of Case Conceptualization

CounselingOutcomesforYouthwithConductDisorders---ARTICLE.pdf

Home >Psychology homework help >Childhood Onset Disorders: Applying the T/C Model of Case Conceptualization

Received 03/02/15 Revised 05/13/15

Accepted 05/19/15 DOI: 10.1002/jcad.12115

Earn CE credit. Visit http://www.prolibraries.com/counseling to purchase and complete the test online.

Conduct disorder (CD) involves the violation of societal norms and/or the basic rights of others as a result of enduring patterns of aggressive behavior (American Psychiatric Asso- ciation [APA], 2013). For children and adolescents through age 19 years, prevalence rates range between 2% and 10% of the population (APA, 2013; Burke, Loeber, & Birmaher, 2002; Erskine et al., 2013; Loeber, Burke, Lahey, Winters, & Zera, 2000). Variation in these rates was accounted for by sample-dependent developmental, cultural, and psychosocial factors such as age, gender, and socioeconomic status. In a nationally representative epidemiological study on lifetime prevalence and age of onset of mental health disorders, Kes- sler et al. (2005) found that 9.5% of the more than 9,000 participants had been diagnosed with CD. This finding reflects the highest prevalence of the three disruptive disorders cur- rently diagnosable in childhood (APA, 2013). It should not be overlooked that epidemiological estimates of CD vary pursuant to substantive modifications of diagnostic criteria in each edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM; Loeber et al., 2000).

Characteristics of CD include verbal and physical hostil- ity and violence toward people or animals, with and without weapons; the deliberate destruction of property via fire setting or other means; lying or stealing; and the disregard of house- hold and/or societal rules, such as making curfew or attending school (APA, 2013). Consequences of CD in the short term include an increased risk for criminal charges, anxiety, de- pression, suicide, substance abuse, peer rejection, relational difficulty, and low academic achievement (Berkout, Young,

Bradley T. Erford, Margaret Ross, and Chelsea Gunther, Education Specialties Department, Loyola University Maryland; Gerta Bardhoshi, Department of Rehabilitation and Counselor Education, University of Iowa; Kelly Duncan, School of Education, Northern State University. Correspondence concerning this article should be addressed to Bradley T. Erford, Education Special- ties Department, Loyola University Maryland, Timonium Graduate Center, 2034 Greenspring Drive, Timonium, MD 21093 (e-mail: [email protected]).

Meta-Analysis of Counseling Outcomes for Youth With Conduct Disorders Bradley T. Erford, Gerta Bardhoshi, Margaret Ross, Chelsea Gunther, and Kelly Duncan

Twenty-one clinical trials were synthesized using a random-effects model, which substantiated that counseling gener- ally produces a medium effect in treating conduct disorder in youth at termination (d+ = 0.30 to 0.57; k = 28). However, the lasting effects at follow-up were unclear because few follow-up studies (k = 13) have been conducted (d+ = –0.53 to 0.58), and only 2 randomized controlled follow-up studies were located. No effects of moderating variables were evident. Implications for counseling practice and outcome research are addressed.

Keywords: meta-analysis, conduct disorder, children, adolescents, counseling

& Gross, 2011; Loeber et al., 2000; Pardini & Fite, 2010). Unfortunately, these outcomes often persist and manifest in adulthood, with poor prognoses and considerable costs to society. Prevalence rates of CD increase between childhood and adolescence, with childhood onset usually involving aggressive boys, and adolescent onset equal across gender, characterized by less physically aggressive behavior, and more likely to involve relational aggression (APA, 2013). Although the majority of individuals with CD remit by adulthood, re- search has consistently found that children with early-onset CD (diagnosed prior to the age of 10 years) are at an increased risk for developing antisocial personality disorder (APD) in adulthood (APA, 2013; Loeber et al., 2000; Pardini & Frick, 2013). This is because childhood-onset individuals with CD tend to be aggressive boys, and because symptoms of CD prior to the age of 18 years are required for an APD diagnosis (APA, 2013). The diagnosis of CD is consistently displayed across race/ethnicity and nationality, although APD is more prevalent among urban environments and lower socioeconomic strata. For individuals to be diagnosed with CD, at least three of 15 DSM-5 (APA, 2013) criteria must be present in the previous 12 months, with at least one of the three present in the previ- ous 6 months. In addition, the child’s behaviors must cause clinically significant impairment to functioning.

Comorbidity between CD and other mental health dis- orders has long been a topic of interest to researchers. In a review of the empirical research on disruptive behavior disor- ders, Loeber et al. (2000) asserted that CD is diagnosed earlier in boys who have attention-deficit/hyperactivity disorder and

Journal of Counseling & Development ■ January 2017 ■ Volume 9536

Erford, Bardhoshi, Ross, Gunther & Duncan

is commonly preceded by a diagnosis of oppositional defi- ant disorder (ODD). CD also co-occurs with anxiety, mood, learning, and substance use disorders (APA, 2013; Berkout et al., 2011; Loeber et al., 2000). Psychosocial factors (e.g., socioeconomic status, parenting, experience with trauma), biological factors (e.g., genetics, prenatal environment, neu- roanatomy, gender), and personal factors (e.g., temperament, intelligence, cognitive functioning) are important risk and treatment factors (Burke et al., 2002). Gender, in particular, has been the cause of much empirical attention in the 21st century, given that males have consistently been diagnosed with disruptive disorders more often than females, especially with regard to early-onset CD (APA, 2013; Berkout et al., 2011; Kessler et al., 2005; Loeber et al., 2000). In addition, CD often manifests differently in females than in males, with females having a more discernible tendency toward internal- izing behaviors, relational aggression, and verbal aggression compared with males.

CD has historically been rather impervious to treatment compared with other childhood disorders (Burke et al., 2002). In light of its prevalence, risk factors, and negative outcomes, it is imperative to understand the strengths and weaknesses of various treatment approaches to develop more effective options. It should be noted that most treatment approaches address symptomatology, because underlying mechanisms of CD are not yet clear (Blair, Leibenluft, & Pine, 2014). Burke et al. (2002) found that preventive approaches targeting specific risk factors were somewhat successful with young children identified as at risk. Along these lines, Farmer, Comp- ton, Burns, and Robertson (2002) reported that participants in parent-training trials in several studies saw better outcomes than control group participants who did not receive parent training. They also noted a dearth in available literature on the efficacy of these programs for parents with children older than 9 years. Outpatient, clinic-based treatments included in their review varied in theory and included psychoanalytic, cognitive behavior, and psychoeducational approaches. All of these interventions reported success, specifically when treat- ment addressed realistic day-to-day situations. Community- based interventions over the past 2 decades have had a mix of positive and negative results, leading to increased caution in the field regarding group work, particularly with delinquent adolescents (Burke et al., 2002; Farmer et al., 2002). Both Burke et al. and Farmer et al. found the most empirical sup- port for multimodal approaches. These approaches, which are based in ecological theory, aim to mediate multiple domains of a child’s life, including school, neighborhood, family, and peer group. The most well-documented and cost-effective of the multimodal approaches was multisystemic therapy, in which therapists intervene with the child one-on-one, in group counseling, and in family therapy to target child-specific risk factors. Given the prevalence of CD in society and the mixed results from clinical trials regarding the short-term

and long-term effectiveness of counseling in the treatment of CD, the purpose of this meta-analysis was to answer two primary questions: (a) Is counseling effective in reducing CD symptoms in school-age youth? and (b) If so, do the effects of counseling last after treatment for CD is terminated?

Method Figure 1 provides a summary of the article selection proce- dures. Counseling in this meta-analysis was defined as any treatment or intervention, provided by a mental health profes- sional or professional-in-training, aimed at the alleviation of conduct symptoms or disorders. We did not include studies of intervention for ODD or another behavior-based disorder un- less those studies also provided treatment for a diagnosed CD.

Inclusion and Exclusion Criteria

The following nine criteria were applied to all clinical trial studies selected for inclusion: (a) Interventions were focused on the reduction of CD symptoms in participants identified with CD; (b) articles were available in English, with no country-of-origin restriction; (c) the study included at least six participants; (d) participants were between the ages of 6 and 18 years; (e) mean difference or mean gain effect-size

Potentially relevant articles (published between 1990 and 2014) identified through computerized searches (j = 685)

FIGURE 1

Flowchart of Included Studies

Note. j = number of articles; k = number of studies/comparisons.

Potentially relevant additional articles identified through search of article reference lists and hand search of prominent journals (j = 48)

Articles excluded after full text review for failure to meet all inclusion criteria (j = 712)

Total number of relevant articles identified and screened for inclusion (j = 733)

Articles finally included in the meta-analysis with usable information (j = 21; k = 28 posttest comparisons [n = 1,733]; k = 15 follow-up comparisons [n = 643]), including: • Wait-list control groups (j = 4; k = 7 posttest comparisons

[n = 570]; k = 0 follow-up comparisons [n = 0]) • Placebo study groups (j = 1; k = 1 posttest comparisons

[n = 91]; k = 1 follow-up comparisons [n = 87]) • Treatment-as-usual comparison groups (j = 10; k = 10

posttest comparisons [n = 771]; k = 1 follow-up comparisons [n = 55])

• Single-study groups (j = 6; k = 10 posttest comparisons [n = 301]; k = 13 follow-up comparisons [n = 501])

Journal of Counseling & Development ■ January 2017 ■ Volume 95 37

Meta-Analysis of Counseling Outcomes for Youth With Conduct Disorders

estimates could be computed from the available data; (f) a standardized instrument was used to measure CD symptoms; (g) individual, group, or family counseling was implemented (drug trials were eliminated); (h) articles appeared in print be- tween 1990 and 2014; and (i) clinical trials used an experimental or quasi-experimental research design, including a single-group or comparison group condition (e.g., wait-list [inactive], pla- cebo [active], or treatment-as-usual [TAU; active] comparison). Sample independence was maintained throughout all analyses. We used publication as a proxy for quality and therefore elimi- nated dissertations, theses, and unpublished manuscripts. We conducted subsequent analyses for publication bias.

Search Strategies

Electronic, reference list, and hand searches of the most relevant journals were used to obtain candidate studies. We searched PsycINFO, Academic Search Premier, and MED- LINE articles from 1990 to 2014 using keywords related to condition (i.e., CD) and intervention (i.e., counseling and psychotherapy). Search parameter restrictions used to identify candidate studies included children and adolescents, English language, peer review, and clinical trials. We then searched reference lists of synthesis articles and selected clinical trials on childhood conduct problems to locate additional candidate studies. Finally, we conducted hand searches of the tables of contents of journals most frequently publishing clinical tri- als on CD (i.e., two or more) from the previous two search procedures (i.e., Journal of the American Academy of Child and Adolescent Psychiatry, Journal of Consulting and Clini- cal Psychology, and Journal of Abnormal Child Psychology).

The title, abstract, and full text of each article were re- viewed, and selection criteria were applied. The third and fourth authors independently determined whether each article met all selection criteria, and the full text of each selected article was submitted for coding and analysis.

Coding Procedures

Selected studies were coded for various participant, design, and method characteristics using a coding manual to allow for subse- quent investigation of moderator or mediator variables in the event that effect-size homogeneity was absent. Participant characteristics included sample size, race/ethnicity, age, gender, completion rate, and nationality. Design characteristics included treatment type, control group type, treatment setting, method of diagnosis, method of recruitment, and use of randomization. Method characteristics included blind assessment, supervision, treatment manual, indi- vidual or group method, number of sessions, duration of sessions, length of study, completion rates, homework, use of therapists specializing in the area of treatment, and characteristics of the counselor/therapist (i.e., degree, discipline, and training level).

These 25 participant, design, and method characteristics were coded and effect sizes computed for each of the 21 included articles. Two judges, each a graduate student who

had previously completed course work in research, statistics, and assessment, and who had completed a training process with practice in coding under supervision of the first author, coded each study independently. The first author mediated coding disagreements between the judges by examining the full text of the article in question and reaching a consensus with the judges. Because of the peer review, randomization, and stringent criteria for study inclusion, a formal evaluation process of study quality was not conducted.

Outcome Measures

The majority of outcome measures used in the 21 included studies were standardized self-report or parent-report mea- sures. Only measures directly assessing conduct problems were included as outcome measures. The most recent version of measures used in more than one study were some form of the Achenbach System of Empirically Based Assessment (e.g., Child Behavior Checklist, Teacher Rating Form, Youth Self-Report; Achenbach & Rescorla, 2001; j = 8, or 38.1% of the 21 included studies), the Revised Problem Behavior Checklist (Quay & Peterson, 1987; j = 5, or 23.8%), the Self- Report Delinquency Checklist (Elliott, Ageton, Huzinga, Knowles, & Canter, 1983; j = 3, or 14.3%), the Eyberg Child Behavior Inventory/Sutter–Eyberg School Behavior Inventory (Eyberg & Pincus, 1999; j = 3, or 14.3%), the Conners–3 Rating Scales (Conners, 2008; j = 2, or 9.5%), and the Parent Daily Report (Chamberlain & Reid, 1987; j = 2, or 9.5%).

Statistical Methods

Only comparable study designs (i.e., all wait-list, all placebo, all TAU, or all single-group designs, independently) with similar effect sizes (e.g., only mean difference effect sizes) were combined (Erford, Savin-Murphy, & Butler, 2010). If studies used multiple outcome measures, we averaged these effect sizes so that a single independent effect size was for- warded for analysis. With regard to follow-up studies, if a study produced several follow-up effect sizes (e.g., 1 year, 2 years), the longest follow-up effect size was forwarded for analysis to give the most conservative follow-up estimate.

Standardized mean difference effect sizes for wait-list, TAU, and placebo group samples were computed using Co- hen’s d. Standardized mean gain effect sizes for single-group samples were calculated using a formula provided by Lipsey and Wilson (2001), with a conservative default value of .70 used in the absence of study-specific reliability values. A positive d indicated that the treatment was effective.

Sample size bias corrections (d′) were made using d′ = d[1 – 3(4N – 9)]. We applied an additional inverse weighting (Erford et al., 2010; Lipsey & Wilson, 2001) to d′ to produce the corrected effect size (d+) prior to combining and averaging study effect sizes for hypothesis testing and homogeneity analyses using the random-effects model. A random-effects model was applied because it assumes that the selected studies were a subset of a

Journal of Counseling & Development ■ January 2017 ■ Volume 9538

Erford, Bardhoshi, Ross, Gunther & Duncan

larger set of studies in existence in the extant literature rather than an all-inclusive finite set. As a result, the random-effects model provides greater external generalizability than a fixed-effects model (Hedges & Olkin, 1985).

In a meta-analysis, two primary hypotheses must be tested for each comparison: (a) Is the mean effect size greater than zero? and (b) Is the distribution of effect sizes homogeneous? To answer the first question, we computed 95% confidence intervals (CIs) for the mean effect sizes (d+; Erford et al., 2010; Lipsey & Wilson, 2001) as a simple assessment to de- termine whether an average effect size was greater than zero. For example, if d+ = 0.15, 95% CI [0.05, 0.25], then the null hypothesis of d+ = 0 can be rejected; the complete range for d+ is greater than zero. However, if d+ = 0.15, 95% CI [–0.05, 0.35], then the null hypothesis is retained because a portion of the 95% CI range is less than or equal to zero.

To answer the second question, we assessed the homogene- ity of the effect-size distributions using Cochran’s Q and an applicable chi-square distribution. If ρ is less than .05, then heterogeneity exists across studies, and the null hypothesis of homogeneity should be rejected. Heterogeneity means that po- tential mediator or moderator variables should be examined using applicable analysis of variance or regression analogs (Hedges & Olkin, 1985; Lipsey & Wilson, 2001). As an additional check on heterogeneity, an inconsistency index (I 2; Higgins, Thompson, Deeks, & Altman, 2003) was computed and interpreted as fol- lows: 0% = no inconsistency, 25% = low inconsistency, 50% = moderate inconsistency, 75% = high inconsistency, and 100% = total inconsistency (total heterogeneity). If I 2 is greater than 50%, significant heterogeneity probably exists and exploration of mediator and moderator variables ensues.

Publication Bias

Multiple methods for assessing publication bias should be under- taken (Beretvas, 2010) in meta-analytic studies, and we subjected each design result to Duval and Tweedie’s (2000; Richardson, Abraham, & Bond, 2012) trim-and-fill procedure, funnel plot analysis, and Rosenthal’s (1979) fail-safe N. Even though the number of studies in each analysis was small, no distributions fell outside of the usual pattern, so publication bias was probably insignificant across the analyses.

Results Computerized and hand searches identified 733 potential can- didate articles, of which 712 were eliminated in subsequent selection reviews of full text because they violated at least one inclusion criterion (see Figure 1). The agreement rate between the two judges was 97% (κ = .94).

Study Characteristics

Of the 21 articles advanced to the coding process (see Figure 1), four (k = 7, where k is the number of studies/comparisons) used

wait-list comparisons, one (k = 1) used a placebo comparison, 10 (k = 10) used a TAU comparison, and six (k = 10) used a single-sample pretest–posttest design. Summary characteristics of these 21 studies (N = 1,733) are provided in Table 1.

Is Counseling Effective With Youth With CD?

We evaluated the treatment effectiveness at termination of CD counseling using wait-list (j = 4, k = 7), placebo (j = 1, k = 1), and TAU (j = 10, k = 10) comparison group studies, and single-group studies (j = 6, k = 10), where j is the number of articles and k is the number of studies/comparisons (some articles presented more than one study or comparison). Mod- ern meta-analytic standards often eliminate studies without control conditions (Higgins & Green, 2011), but because so few randomized controlled clinical trials were located in the literature (i.e., 15) we retained the six single-group articles yielding 10 viable studies as supplemental analyses. When interpreting d, Lipsey and Wilson (2001) suggested that 0 means no effect, less than .30 is a small effect, .50 is a medium effect, and greater than .67 is a large effect. Furthermore, because d is actually a specialized z score, it can be directly transformed into a percentile rank and used to reveal how much improvement the average participant in the treatment condition made compared with the average participant in the comparison group (Erford et al., 2010).

Wait-list comparison groups. Seven comparisons (n = 570) across four studies (Kumar, 2009; McNeil, Eyberg, Eisenstadt, Newcomb, & Funderburk, 1991; Sanders, Markie-Dadds, Tully, & Bor, 2000; van Manen, Prins, & Emmelkamp, 2004) yielded an average corrected effect size (d+) of 0.53, 95% CI [0.35, 0.71]. Because the entire 95% CI range is greater than zero, the null hypothesis of d+ = 0 can be rejected, and it can be concluded that counseling was effective at termination. A d+ of 0.53 is a medium effect and means that the average participant at termination scored at the 70th percentile of the control group’s score distribu- tion. In a seven-study analysis, a d+ of 0.53 has a fail-safe N of 364, meaning that 364 wait-list studies would need to be located with an effect size of zero to mitigate the d+ of 0.53 down to 0.01. The funnel plot showed no outliers, and the trim-and-fill procedure resulted in ρ greater than .05. These three analyses indicated a very low probability of publication bias. Results indicated that the test for homogene- ity of these seven comparisons was statistically nonsignificant, Cochran’s Q(6) = 5.30, ρ > .05, meaning that the null hypothesis for homogeneity was retained and the set of seven effect sizes was homogeneous and consistent. To corroborate this result, we computed I2, which, at 0%, also indicated homogeneity (i.e., I2 < 50%). Because the distribution of effect sizes was homogeneous, there was no need to conduct moderator or mediator analyses.

Placebo comparison groups. The single mean difference comparison for the single article (Rohde, Clarke, Mace,

Journal of Counseling & Development ■ January 2017 ■ Volume 95 39

Meta-Analysis of Counseling Outcomes for Youth With Conduct Disorders

Jorgensen, & Seeley, 2004) reporting a placebo comparison (n = 91) yielded a d+ equal to 0.30, 95% CI [–0.11, 0.71], which was not significantly greater than zero. The average participant in the treatment condition performed at the 62nd percentile of the placebo comparison group’s score distribution, indicating no treatment effect. Analysis resulted in a fail-safe N of only 29 studies. Because there was only one effect size, no funnel plot or trim-and-fill procedure was conducted, and homogeneity was evident by default.

TAU comparison groups. Ten comparisons from the 10 ar- ticles (Apsche, Bass, Zeiter, & Houston, 2008; Bank, Marlowe, Reid, Patterson, & Weinrott, 1991; Borduin et al., 1995; Cham- berlain & Reid, 1998; Henggeler, Melton, Brondino, Scherer, & Hanley, 1997; Henggeler, Melton, & Smith, 1992; Kendall, Reber, McLeer, Epps, & Ronan, 1990; Scherer, Brondino, Henggeler, Melton, & Hanley, 1994; Vitaro & Tremblay, 1994; Weisz et al., 2012) reporting TAU designs (n = 771) combined for a d+ equal to 0.55, 95% CI [0.40, 0.70], a result that was

Note. unk = unknown; TAU = treatment as usual; CBT = cognitive behavior therapy; REBT = rational-emotive behavior therapy; ODD = op- positional defiant disorder. aMean age in years.

TABLE 1

Characteristics of Individual Studies Used in the Conduct Disorder (CD) Meta-Analysis

Apsche et al. (2008)

Bank et al. (1991) Borduin et al.

(1995) Chamberlain &

Reid (1998) Dadds & McHugh

(1992) Henggeler et al.

(1992)

Henggeler et al. (1997)

Kazdin et al. (1992)

Kendall et al. (1990)

Kumar (2009)

McNeil et al. (1991)

Rohde et al. (2004)

Sanders et al. (2000)

Scherer et al. (1994)

Schneider (1991)

Stadler et al. (2008)

van Manen et al. (2004)

Vitaro & Tremblay (1994)

Webster-Stratton (1994)

Weisz et al. (2012)

Whitmore et al. (2000)

Study Agean % MaleSummary unk

14.0 14.8

14.9

4.5

15.2

10.3

10.7

unk

4.7

unk

15.1

10.5

10.4

11.2

6.1

5.0

10.6

15.5

55 126

140

200

104

174

unk

100 68

100

unk

100

Family mode deactivation therapy

Parent-training interventions Multisystemic treatment

Multidimensional foster care

Child management and ally support training

Multisystemic therapy

Cognitive problem-solving skills and parent manage- ment training

CBT

REBT

Parent–child interaction therapy

CBT with comorbid CD and depression

Comparing enhanced, standard, and self-directed behavioral family interventions

Multisystemic family preservation therapy

Social skills training and desensitization strategies

CBT

Social skills training and social- cognitive group treatment

Prevention of aggression program

Videotape and live parent training

Modular and standard condi- tion

Outpatient treatment

Conduct Outcome Measure% White Control

Group Type Child Behavior Checklist–Externalizing

Total offenses Revised Problem Behavior Checklist

Elliott Behavior Checklist; runaway; days in lockup

Revised Problem Behavior Checklist

Revised Problem Behavior Checklist; Self-Report Delinquency Checklist; Missouri Peer Relationship Inventory

Revised Problem Behavior Checklist; Self-Report Delinquency Checklist

Child Behavior Checklist; Teacher Rating Form; Parent Daily Report; Interview for Antisocial Behavior; Self- Report Delinquency Checklist

Child Behavior Checklist–Externalizing; Conners Teacher Questionnaire

Youth Self-Report Conduct Problems subscale

Revised Conners Teacher Rating Scale; Sutter–Eyberg School Behavior Inventory

Child Behavior Checklist–Externalizing

Eyberg Child Behavior Inventory; Par- ent Daily Report; Parent Problem Checklist

Revised Problem Behavior Checklist

Observed aggression

Child Behavior Checklist; number of CD/ ODD symptoms

Child Behavior Checklist

Social Behavior Questionnaire

Child Behavior Checklist; Eyberg Child Behavior Inventory

Brief Problem Checklist; Top Problems Assessment

CD symptoms

0 70

unk

100

unk

TAU

TAU TAU

TAU

single

TAU

single

TAU

wait list

placebo

wait list

TAU

single

wait list

TAU

single

TAU

single

Journal of Counseling & Development ■ January 2017 ■ Volume 9540

Erford, Bardhoshi, Ross, Gunther & Duncan

significantly greater than zero. Analysis resulted in a fail-safe N of 540 studies, a funnel plot analysis with no outliers, and a trim-and-fill procedure with no appreciable effect. Thus, the average participant in the treatment condition performed at the 71st percentile of the TAU comparison group’s score dis- tribution, indicating a small-to-large treatment effect. Again, the distribution of effect sizes was very homogeneous, Q(9) = 7.43, ρ > .05, with an I 2 of 0%.

Single-group studies. Across six articles, 10 single-group comparison studies (n = 301) were reported (Dadds & McHugh, 1992; Kazdin, Siegel, & Bass, 1992; Schneider, 1991; Stadler et al., 2008; Webster-Stratton, 1994; Whit- more, Mikulich, Ehlers, & Crowley, 2000), yielding a d+ of 0.57, 95% CI [0.47, 0.67]. This d+ is greater than zero and has a fail-safe N of 280 studies, with funnel plot and trim-and-fill results that are within normal limits. Thus, these results are quite robust. The homogeneity analysis indicated substantial homogeneity, Q(4) = 10.41, ρ > .05, which was corroborated by an I 2 of 14%. A d+ of 0.57 is a medium effect, indicating the average treatment group participant at termination displayed fewer conduct problems than did 72% of the pretest participants.

In summary, all except the placebo condition (k = 1) yielded weighted effect sizes (d+) significantly greater than zero at termination, meaning that the treatments for CD were effective and that the average treatment participant was be- tween the 62nd and 72nd percentile of the comparison group’s score distribution at termination. Furthermore, Cochran’s Q and I 2 for each analysis indicated significant homogeneity, and therefore no effects of moderating or mediating variables. Table 2 provides a summary of these effect-size statistics.

Do the Effects of Counseling Last for Youth With CD?

To determine the effectiveness of counseling after termina- tion, we used the follow-up point furthest from termination to yield the most conservative estimate. No wait-list follow-up studies were located.

Placebo comparison group studies. One sample from one placebo study (Rohde et al., 2004) provided follow-up data (n = 87), resulting in a d+ of –0.53, 95% CI [–0.96, –0.10], which is not greater than zero. Indeed, the d+ was negative, meaning that the average treatment group member performed at the 30th percentile and was worse off than the average pla- cebo condition participant. A single trial means that no funnel plot, trim-and-fill, or homogeneity analysis was necessary. These results should be interpreted with caution because of the single study involved.

TAU comparison group studies. The effects of counsel- ing in the treatment of CD in school-age children at longest follow-up for the single study (Vitaro & Tremblay, 1994) providing TAU follow-up data (n = 55) resulted in a d+ of 0.08, 95% CI [–0.45, 0.61]. Because this result was not greater than zero, the null hypothesis was retained. The fail-safe N was only seven studies. Funnel plot, trim-and-fill, and homogene- ity procedures were not needed because of the single-study analysis. A d+ of 0.08 indicates no effect of treatment and means that the average participant in the treatment condition displayed fewer conduct problems than did approximately 53% of the TAU group participants at follow-up.

Single-group studies. Several TAU and wait-list studies provided single-group follow-up studies because the TAU control conditions received the treatments immediately after termination of the study. Thus, k = 13 samples provided follow-up data (n = 501), yielding a d+ of 0.58, 95% CI [0.50, 0.66], which was greater than zero. The fail-safe N was 741 studies, and the funnel plot analysis and trim-and-fill procedure yielded nonsignificant results. A d+ of 0.58 is a medium effect, indicating that the average participant in the treatment condition at follow-up displayed fewer conduct problems than did approximately 72% of the pretest partici- pants. This distribution of 13 effect sizes was homogeneous, Q(12) = 15.58, ρ > .05, with an I 2 of 23%. Therefore, no analysis for moderator or mediator variables was conducted. Some evidence exists (see Table 3) that treatment gains were maintained beyond 2 years after termination, given that we

TABLE 2

Summary of Posttest and Follow-Up Results

Condition

Termination (posttest) results Wait list Placebo Treatment as usual Single group

Longest interval follow-up results Wait list Placebo Treatment as usual Single group

570 91 771 301

87 55 501

Note: j = number of articles; k = number of studies/comparisons. ad+ > 0 (p < .05).

.53a .30 .55a .57a

–.53 .08 .58a

[0.35, 0.71] [–0.11, 0.71] [0.40, 0.70] [0.47, 0.67]

[–0.96, –0.10] [–0.45, 0.61] [0.50, 0.66]

ρ > .05

ρ > .05 ρ > .05

ρ > .05

70 62 71 72

30 53 72

Q95% CId+n Percentile

7 1 10 10

0 1 1 13

4 1 10 6

0 1 1 6

Journal of Counseling & Development ■ January 2017 ■ Volume 95 41

Meta-Analysis of Counseling Outcomes for Youth With Conduct Disorders

observed the following progression for the studies reporting follow-up effects: posttest d+ = 0.58 (k = 13), 1- to 6-month follow-up d+ = 0.81 (k = 2), 7- to 12-month follow-up d+ = 1.18 (k = 2), and 2-year follow-up d+ = 0.53 (k = 9). Again, these results should be interpreted with caution because of the small number of studies involved and the fact that different studies composed the different follow-up intervals.

In summary, only the single-group conditions resulted in an average weighted effect size (d+) that was significantly greater than zero at follow-up intervals, whereas the d+ values of the placebo and TAU comparisons were not significantly greater than zero. Because we were unable to locate any wait-list follow-up studies and given the low power resulting from a small sample analysis, results regarding the long-term treatment effectiveness of counseling on CD should be interpreted with caution. However, single-group treatment gains appeared to have been maintained up to 2 years after termination, with the average participant in the single-group treatment condition at follow-up standing at the 72nd percentile of the pretest group’s score distribution. All tests of homogeneity (Cochran’s Q and I2) suggested significant homogeneity, and therefore no effects of moderating or mediating variables. Table 3 provides a sum- mary of the follow-up effect-size statistics.

Discussion Given the poor prognosis and potential for long-term and per- sistent psychopathology in youth with CD (Blair et al., 2014), it is imperative that evidence-based practices be available for practitioners working with this population. The studies contained in this meta-analysis used a variety of treatment modalities, including parent training, multisystemic therapy, individual approaches, and group therapy, with effectiveness at termination being evaluated using wait-list, placebo, and TAU comparisons, and single-group studies. A random- effects model synthesizing effect sizes from 21 clinical trials published between 1990 and 2014 produced mostly medium effect-size statistics, with the average participant displaying fewer conduct symptoms than 62% to 72% of control group participants at termination. The results of study comparisons at termination lend evidence that counseling is effective in treating CD in youth.

Very few controlled, follow-up studies with this population have been undertaken to date; thus, it is unclear whether the results of counseling can be viewed as long lasting in treating CD. We located a single placebo, single TAU, and 13 single- group comparisons to evaluate the treatment effectiveness furthest from termination, aiming for a conservative estimate. An analysis of 13 single-group follow-up comparisons rang- ing from 6 months to beyond 2 years produced evidence of a medium effect of treatment, with gains potentially being maintained beyond 2 years. Indeed, the average participant in the treatment group displayed less symptomatology at the longest interval follow-up than did approximately 72% of par- ticipants at pretest. However, this conclusion should be viewed as tentative because of the low number of follow-up studies.

Selected studies also included a wide range of therapeutic approaches (e.g., cognitive behavior therapy, multisystemic therapy) and modalities (i.e., individual, group, or family methods). Because the purpose of this meta-analysis was to determine the effectiveness and potential staying power of counseling in reducing CD symptomatology in school-age youth, questions regarding the advantage of one approach or modality over another may arise. However, our analysis indicated homogeneity of the effect-size distributions, thereby rejecting the existence of potential mediator or moderator variables that could affect treatment effectiveness in this population. Therefore, no moderating effects were noted for any of the 25 coded study characteristics, including client, counselor, and treatment modality factors. The treatment effectiveness of counseling remained consistent across mul- tiple study comparisons with diverse participant, design, and method characteristics, a notably robust result.

The overall conclusions of this study regarding treatment effectiveness at termination and the potential for lasting ef- fects at follow-up can be applied across relevant populations and treatment variations. Reported mean ages of clients across studies ranged from 4.5 to 15.5 years, and client age along with gender and race/ethnicity did not appear to influence the effectiveness of counseling. The most frequently used treatment interventions with school-age youth among the 21 studies included in this meta-analysis appeared to be multi- systemic therapy and cognitive behavior therapy, which is reflective of the larger literature, with parent training, family therapy, and social skills training following suit. Similarly, the length of treatment, the use of a treatment manual, or the use of a therapist specializing in the area of treatment did not result in significant differences in effect sizes, further strengthening the generalizability of these results to a wide range of practices and clinicians.

Limitations of This Meta-Analysis

Stringent methodological practices were used in conducting this meta-analysis. They included (a) nine inclusion criteria, (b) comprehensive literature search strategies, (c) use of

TABLE 3

Follow-Up Studies Continuum

Condition

Wait list Placebo Treatment as usual Single group

Note. M d+

= mean effect-size estimate. ak = 1. bk = 2. ck = 9.

None None None 0.81b

None –0.53a None 1.18b

None None 0.08a 0.53c

7–12 Months M

<6 Months M

24+ Months M

Journal of Counseling & Development ■ January 2017 ■ Volume 9542

Erford, Bardhoshi, Ross, Gunther & Duncan

standardized instruments to measure conduct symptoms, (d) random-effects modeling, (e) sample size bias correc- tion, (f) multiple methods for assessing publication bias and homogeneity, and (g) inverse variance weighting procedures. Although the rigorous criteria for inclusion enhanced the quality of this meta-analysis, it is possible that some viable research studies with signif icant results were excluded. Indeed, the small number of articles that met the inclusion criteria is possibly the most significant limitation of this study. However, despite a limited number of studies, using a conservative methodology led to estimates that are more reflective of real-life outcomes in counseling, which will better inform clinicians than results obtained from including less rigorous studies.

Because this meta-analysis included only 21 candidate published articles and very few randomized controlled clinical trials were located, we decided to include six single-group articles as a supplemental analysis, which yielded 10 follow- up comparisons. The inclusion of studies without control conditions is a potential limitation of this meta-analysis. Unfortunately, only two randomized controlled follow-up trials were located, so the noncontrolled, single-group, follow- up studies are the best evidence of the long-term effects of counseling at this time. In addition, many of the follow-up study comparisons evaluating the long-lasting effects of treatment were from TAU and wait-list trials that resulted in single-group follow-up studies, given that control group par- ticipants received the treatment immediately after termination. Although this practice is ethically and clinically justifiable, it did limit the number of available comparisons at follow-up intervals. Finally, we could locate only one study using pla- cebo comparisons for posttreatment or follow-up, one study using a TAU comparison for follow-up, and no studies using wait-list comparisons for follow-up, thus leading to limita- tions in terms of stability and sufficient power of the respective analyses. In short, a primary limitation of this meta-analysis is the small number of clinical trials studying the immediate and long-term effectiveness of counseling youth with CD.

Implications for Counseling Practice

The present study provides evidence that counseling produces a medium effect in treating school-age youth diagnosed with CD. Indeed, counseling appears to be quite effective whether the approach is individual or group, and irrespective of treat- ment setting or the specific therapeutic approach applied. This finding provides counselors with opportunities to make treatment decisions based on client needs, available resources, and cost considerations, without having to sacrifice the quality or effectiveness of the intervention with school-age youth.

More information is needed on how counseling profes- sionals can best extend effectiveness after treatment ends to improve longer term results (APA, 2013). It is possible that continuing treatment by extending session intervals or

implementing posttreatment booster sessions could help maintain treatment effects over the long term (Erford et al., 2011, 2013). More clinical trials examining the effec- tiveness of counseling in various follow-up phases will help clarify the progression of treatment effects and the potential optimal stages of intervention to maintain significant gains. Using booster sessions with children and adolescents who have already received treatment for CD at optimal postintervention intervals and based on evaluated treatment schedules could further improve the practice of counseling with this population.

The school-age years are an extremely important time period in a child’s development (Broderick & Blewitt, 2014). Although focusing on the effectiveness and staying power of counseling in treating CD in children is important, an emphasis should also be placed on the prevention and early intervention of mental health disorders (APA, 2013; Burke et al., 2002; Farmer et al., 2002). Given the dismal prognosis for early-onset CD and the increased risk of diagnosis of APD in adulthood (APA, 2013), clinicians who understand and screen for risk factors and periodically assess function- ing at school and home may better mobilize resources and available interventions.

Implications for Counseling Research

Intensive early intervention is key, and published studies with comprehensive assessment practices are needed to identify and evaluate the effectiveness of CD treatment in youth. The majority of outcome measures used in the included studies were either self-report or parent-report measures. Inclusion of family members, caregivers, and school personnel is vital in implementing a multisystemic approach, and future research evaluating treatment effectiveness should be grounded in a multi-informant approach. Although 10 of the 21 identified studies used more than one outcome measure to evaluate treatment in youth diagnosed with CD, only one study used a combination of self-report, parent, teacher, and counselor evaluation to determine treatment outcome. Future research should consistently use assessment best practices by triangu- lating reports from parents, teachers, and clients along with clinical judgment.

More studies are needed that adequately examine the lasting effects of counseling by using follow-up comparisons at multiple intervals. Because of the serious long-lasting problems that youth with CD may experience even as adults, studies of the effects of interventions beyond termination are needed. Designing and conducting outcome studies that adequately quantify interventions and examine treatment effects beyond 2 years will not only allow for rigorous evalu- ation of services, but also provide much-needed information regarding the prognosis of this hard-to-treat disorder and opportunities for replicating and enhancing effectiveness. It is imperative that counseling professionals working with this population identify evidence-based treatments that reduce CD

Journal of Counseling & Development ■ January 2017 ■ Volume 95 43

Meta-Analysis of Counseling Outcomes for Youth With Conduct Disorders

symptomatology in children and adolescents and continue exploring opportunities for innovation and improved, long- lasting outcomes.

References References marked with an asterisk indicate studies included in the

meta-analysis. Achenbach, T. M., & Rescorla, L. A. (2001). Manual for the ASEBA

School-Age Forms & Profiles. Burlington: University of Vermont, Research Center for Children, Youth, & Families.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: Author.

*Apsche, J. A., Bass, C. K., Zeiter, J. S., & Houston, M. A. (2008). Family mode deactivation therapy in a residential setting: Treating adolescents with conduct disorder and multi-axial diagnosis. International Journal of Behavioral Consultation and Therapy, 4, 328–339.

*Bank, L., Marlowe, J. H., Reid, J. B., Patterson, G. R., & Weinrott, M. R. (1991). A comparative evaluation of parent-training inter- ventions for families of chronic delinquents. Journal of Abnormal Child Psychology, 19, 15–33. doi:10.1007/BF00910562

Beretvas, S. N. (2010). Meta-analysis. In G. R. Hancock & R. O. Mueller (Eds.), The reviewer’s guide to quantitative methods in the social sciences (pp. 255–263). New York, NY: Routledge.

Berkout, O. V., Young, J. N., & Gross, A. M. (2011). Mean girls and bad boys: Recent research on gender differences in con- duct disorder. Aggression and Violent Behavior, 16, 503–511. doi:10.1016/j.avb.2011.06.001

Blair, R. J. R., Leibenluft, E., & Pine, D. S. (2014). Conduct disorder and callous–unemotional traits in youth. The New England Jour- nal of Medicine, 371, 2207–2216. doi:10.1056/NEJMra1315612

*Borduin, C. M., Mann, B. J., Cone, L. T., Henggeler, S. W., Fucci, B. R., Blaske, D. M., & Williams, R. A. (1995). Multisystemic treatment of serious juvenile offenders: Long-term prevention of criminality and violence. Journal of Consulting and Clinical Psychology, 63, 569–578. doi:10.1037/0022-006X.63.4.569

Broderick, P. C., & Blewitt, P. (2014). The life span: Human devel- opment for helping professionals (4th ed.). Upper Saddle River, NJ: Pearson Merrill.

Burke, J. D., Loeber, R., & Birmaher, B. (2002). Oppositional defiant and conduct disorder: A review of the past 10 years, Part II. Jour- nal of the American Academy of Child and Adolescent Psychiatry, 41, 1275–1293. doi:10.1097/01.CHI.0000024839.60748.E8

Chamberlain, P., & Reid, J. B. (1987). Parent observation and report of child symptoms. Behavioral Assessment, 9, 97–109.

*Chamberlain, P., & Reid, J. B. (1998). Comparison of two commu- nity alternatives to incarceration for chronic juvenile offenders. Journal of Consulting and Clinical Psychology, 66, 624–633. doi:10.1037/0022-006X.66.4.624

Conners, C. K. (2008). Manual for the Conners–3. North Tonawanda, NY: Multi-Health Systems.

*Dadds, M. R., & McHugh, T. A. (1992). Social support and treat- ment outcome in behavioral family therapy for child conduct problems. Journal of Consulting and Clinical Psychology, 60, 252–259. doi:10.1037/0022-006X.60.2.252

Duval, S., & Tweedie, R. (2000). Trim and fill: A simple funnel- plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics, 56, 455–463. doi:10.1111/j.0006- 341X.2000.00455.x

Elliott, D. S., Ageton, S. S., Huzinga, D., Knowles, B. A., & Canter, R. J. (1983). The prevalence and incidence of delinquent behav- ior: 1976–1980 (National Youth Survey Report No. 26). Boulder, CO: Behavioral Research Institute.

Erford, B. T., Erford, B. M., Lattanzi, G., Weller, J., Schein, H., Wolf, E., . . . Peacock, E. (2011). Counseling outcomes from 1990 to 2008 for school-age youth with depression: A meta- analysis. Journal of Counseling & Development, 89, 439–457. doi:10.1002/j.1556-6676.2011.tb02841.x

Erford, B. T., Richards, T., Peacock, E. R., Voith, K., McGair, H., Muller, B., . . . Chang, C. Y. (2013). Counseling and guided self- help outcomes for clients with bulimia nervosa: A meta-analysis of clinical trials from 1980 to 2010. Journal of Counseling & De- velopment, 91, 152–172. doi:10.1002/j.1556-6676.2013.00083.x

Erford, B. T., Savin-Murphy, J. A., & Butler, C. (2010). Conducting a meta-analysis of counseling outcome research: Twelve steps and practical procedures. Counseling Outcome Research and Evaluation, 1, 19–43. doi:10.1177/2150137809356682

Erskine, H. E., Ferrari, A. J., Nelson, P., Polanczyk, G. V., Flaxman, A. D., Vos, T., . . . Scott, J. G. (2013). Research review: Epidemio- logical modelling of attention-deficit/hyperactivity disorder and conduct disorder for the Global Burden of Disease Study 2010. Journal of Child Psychology and Psychiatry, 54, 1263–1274. doi:10.1111/jcpp.12144

Eyberg, S., & Pincus, D. (1999). Eyberg Child Behavior Inventory & Sutter-Eyberg Student Behavior Inventory–Revised: Professional manual. Odessa, FL: Psychological Assessment Resources.

Farmer, E. M. Z., Compton, S. N., Burns, J. B., & Robertson, E. (2002). Review of the evidence base for treatment of childhood psychopathology: Externalizing disorders. Journal of Consulting and Clinical Psychology, 70, 1267–1302. doi:10.1037/0022- 006X.70.6.1267

Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta- analysis. Orlando, FL: Academic Press.

*Henggeler, S. W., Melton, G. B., Brondino, M. J., Scherer, D. G., & Hanley, J. H. (1997). Multisystemic therapy with violent and chronic juvenile offenders and their families: The role of treat- ment fidelity in successful dissemination. Journal of Consulting and Clinical Psychology, 65, 821–833. doi:10.1037/0022- 006X.65.5.821

*Henggeler, S. W., Melton, G. B., & Smith, L. A. (1992). Family preservation using multisystemic therapy: An effective alternative to incarcerating serious juvenile offenders. Journal of Consult- ing and Clinical Psychology, 60, 953–961. doi:10.1037/0022- 006X.60.6.953

Journal of Counseling & Development ■ January 2017 ■ Volume 9544

Erford, Bardhoshi, Ross, Gunther & Duncan

Higgins, J. P. T., & Green, S. (Eds.). (2011). Cochrane handbook for systematic reviews of interventions (Version 5.1.0). Retrieved from http://handbook.cochrane.org/

Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analyses. British Medical Journal, 327, 557–560. doi:10.1136/bmj.327.7414.557

*Kazdin, A. E., Siegel, T. C., & Bass, D. (1992). Cognitive problem- solving skills training and parent management training in the treatment of antisocial behavior in children. Journal of Consult- ing and Clinical Psychology, 60, 733–747. doi:10.1037/0022- 006X.60.5.733

*Kendall, P. C., Reber, M., McLeer, S., Epps, J., & Ronan, K. R. (1990). Cognitive-behavioral treatment of conduct-disordered children. Cognitive Therapy and Research, 14, 279–297. doi:10.1007/BF01183997

Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62, 593–602. doi:10.1001/archpsyc.62.6.593

*Kumar, G. V. (2009). Impact of rational-emotive behaviour therapy (REBT) on adolescents with conduct disorder (CD). Journal of the Indian Academy of Applied Psychology, 35, 103–111.

Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.

Loeber, R., Burke, J. D., Lahey, B. B., Winters, A., & Zera, M. (2000). Oppositional defiant and conduct disorder: A review of the past 10 years, Part I. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 1468–1484. doi:10.1097/00004583- 200012000-00007

*McNeil, C. B., Eyberg, S., Eisenstadt, T. H., Newcomb, K., & Funderburk, B. (1991). Parent–child interaction therapy with behavior problem children: Generalization of treatment effects to the school setting. Journal of Clinical Child Psychology, 20, 140–151. doi:10.1207/s15374424jccp2002_5

Pardini, D. A., & Fite, P. J. (2010). Symptoms of conduct disorder, oppositional defiant disorder, attention-deficit/hyperactivity disorder, and callous–unemotional traits as unique predictors of psychosocial maladjustment in boys: Advancing evidence base for DSM-V. Journal of the American Academy of Child and Adolescent Psychiatry, 49, 1134–1144. doi:10.1016/j. jaac.2010.07.010

Pardini, D., & Frick, P. J. (2013). Multiple developmental pathways to conduct disorder: Current conceptualizations and clinical implications. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 22, 20–25.

Quay, H. C., & Peterson, D. R. (1987). Manual for the Revised Behav- ior Problem Checklist. Coral Gables, FL: University of Miami.

Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university students’ academic performance: A sys- tematic review and meta-analysis. Psychological Bulletin, 138, 353–387. doi:10.1037/a0026838

*Rohde, P., Clarke, G. N., Mace, D. E., Jorgensen, J. S., & Seeley, J. R. (2004). An efficacy/effectiveness study of cognitive-behavioral treatment for adolescents with comorbid major depression and conduct disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 660–668. doi:10.1097/01. chi.0000121067.29744.41

Rosenthal, R. (1979). The file drawer problem and tolerance for null results. Psychological Bulletin, 86, 638–641. doi:10.1037/0033- 2909.86.3.638

*Sanders, M. R., Markie-Dadds, C., Tully, L. A., & Bor, W. (2000). The Triple P-Positive Parenting Program: A comparison of enhanced, standard, and self-directed behavioral family interven- tion for parents of children with early onset conduct problems. Journal of Consulting and Clinical Psychology, 68, 624–640. doi:10.1037/0022-006X.68.4.624

*Scherer, D. G., Brondino, M. J., Henggeler, S. W., Melton, G. B., & Hanley, J. H. (1994). Multisystemic family preservation therapy: Preliminary findings from a study of rural and minority serious adolescent offenders. Journal of Emotional and Behavioral Dis- orders, 2, 198–206. doi:10.1177/106342669400200402

*Schneider, B. H. (1991). A comparison of skill-building and desensitization strategies for intervention with aggressive children. Aggressive Behavior, 17, 301–311. doi:10.1002/1098- 2337(1991)17:6<301::AID-AB2480170602>3.0.CO;2-8

*Stadler, C., Grasmann, D., Fegert, J. M., Holtmann, M., Poustka, F., & Schmeck, K. (2008). Heart rate and treatment effect in children with disruptive behavior disorders. Child Psychiatry and Human Development, 39, 299–309. doi:10.1007/s10578-007-0089-y

*van Manen, T. G., Prins, P. J. M., & Emmelkamp, P. M. G. (2004). Reducing aggressive behavior in boys with a so- cial cognitive group treatment: Results of a randomized, controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 1478–1487. doi:10.1097/01. chi.0000142669.36815.3e

*Vitaro, F., & Tremblay, R. E. (1994). Impact of a prevention program on aggressive children’s friendships and social adjust- ment. Journal of Abnormal Child Psychology, 22, 457–475. doi:10.1007/BF02168085

*Webster-Stratton, C. (1994). Advancing videotape parent training: A comparison study. Journal of Consulting and Clinical Psychol- ogy, 62, 583–593. doi:10.1037/0022-006X.62.3.583

*Weisz, J. R., Chorpita, B. F., Palinkas, L. A., Schoenwald, S. K., Miranda, J., Bearman, S. K., . . . Gibbons, R. D. (2012). Testing standard and modular designs for psychotherapy treating depres- sion, anxiety, and conduct problems in youth: A randomized effectiveness trial. Archives of General Psychiatry, 69, 274–282. doi:10.1001/archgenpsychiatry.2011.147

*Whitmore, E. A., Mikulich, S. K., Ehlers, K. M., & Crowley, T. J. (2000). One-year outcome of adolescent females referred for conduct disorder and substance abuse/dependence. Drug and Alcohol Dependence, 59, 131–141. doi:10.1016/S0376- 8716(99)00112-X

Copyright of Journal of Counseling & Development is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.